- Email: [email protected]
Isolation of iron-oxidizing bacteria from corroded concretes of sewage treatment plants
JOURNAL OF BIOSCIENCEAND BIOENGINEERING Vol. 88, No. 3, 300-305. 1999
Isolation of Iron-Oxidizing Bacteria from Corroded Concretes of SewageTreatment Plants TERUNOBU
MAEDA,’ ATSUNORI NEGISHI,’ HIROMI KOMOT0,2 YUKO OSHIMA,2 KAZUO KAMIMURA,2 AND TSUYOSHI SUGI02* Technical Research Institute, Hazama Corporation, 515-l Nishimukai, Karima, Tsukuba 305-0822’ and Department of Biological Function and Genetic Resources Science, Faculty of Agriculture, Okayama University, I-I-I Tsushima Naka, Okayama 700-8530,2 Japan Received 10 March 1999/Accepted 10 June 1999
Thirty-six strains of iron-oxidizing bacteria were isolated from corroded concrete samples obtained at eight sewage treatment plants in Japan. All of the strains isolated grew autotrophically in ferrous sulfate (3.0%), elemental sulfur (1.0%) and FeS (1.0%) media (pH 1.5). Washed intact cells of the 36 isolates had activities to oxidize both ferrous iron and elemental sulfur. Strain SNA-5, a representative of the isolated strains, was a gram-negative, rod-shaped bacterium (OS-o.6 X 0.9-1.5 q). The mean G+C content of its DNA was 55.9 mol%. The pH and temperature optima for growth were 1.5 and 30% and the bacterium had activity to assimilate ‘4c02 into the cells when ferrous iron or elemental sulfur was used as a sole source of energy. These results suggest that SNAQ is Thiobacillus ferrooxidans strain. The pHs and numbers of iron-oxidizing bacteria in corroded concrete samples obtained by boring to depths of O-l, 1-3, and 3-5 cm below the concrete surface were respectively 1.4, 1.7, and 2.0, and 1.2 x 108, 5 X lo’, and 5 X lo6 cells/g concrete. The degree of corrosion in the sample obtained nearest to the surface was more severe than in the deeper samples. The findings indicated that the levels of acidification and corrosion of the concrete structure corresponded with the number of iron-oxidizing bacteria in a concrete sample. Sulfuric acid produced by the chemolithoautotrophic sulfuroxidizing bacterium Thiobacillus thiooxidansis known to induce concrete corrosion. Since not only T. thiooxidans but also T. ferrooxidans can oxidize reduced sulfur compounds and produce sulfuric acid, the results strongly suggest that T. ferrooxidans as well as T. thiooxidans is involved in concrete corrosion. [Key words: corrosion]
iron-oxidizing
bacterium,
ThiobaciNus ferrooxidans, sulfur-oxidizing
Parker first presented evidence that concrete structures are corroded by sulfuric acid produced by the sulfuroxidizing bacterium Thiobacillus concretivourus (l-3). Concrete in its main structure is composed of calcium oxides, hydroxides, and carbonates, and these compounds can react with sulfuric acid to give calcium sulfate (gypsum) (4). This resultant calcium loss greatly diminishes the strength of concrete structures. The bacterium isolated by Parker was later identified as Thiobacillus thiooxidans, and thiobacilli other than T. thiooxidans (5, 6), namely Thiobacillus neapolitanus, T. in termed&, Thiobacillus novellus, and Thiobacillus versutus have also been isolated from sewer systems and corroded concrete (7-10). The iron-oxidizing bacterium Thiobacillus ferrooxidans mainly inhabits drainage areas in acid mines and plays a crucial role in the bacterial leaching of sulfide ores (11-15). As well as T. thiooxidans, T. ferrooxidans is also known to oxidize reduced inorganic sulfur compounds and produce sulfuric acid. The enzymes involved in the oxidation of inorganic sulfur compounds have been purified from the bacterium (16-19). The ability of T. ferrooxidans to produce sulfuric acid suggested that in addition to T. thiooxidans, T. ferrooxidans strains are also likely to be present in corroded concrete and play a role in concrete corrosion. We were able to confirm that this is the case, as well as to clarify the mechanism of concrete corrosion more precisely, by isolating iron-oxidizing bacteria from corroded concrete samples obtained from eight sewage treatment plants in * Corresponding
bacterium,
concrete,
Japan. To our knowledge, there have been no previous reports on concrete corrosion by microorganisms that have focussed on iron-oxidizing bacteria. MATERIALS
AND METHODS
Microorganisms, media, conditions of cultivation, and isolation procedures T. ferrooxidans AP19-3 (16, 19) and 36 strains of iron-oxidizing bacteria isolated from corroded concrete samples obtained at eight sewage treatment plants in Japan were used throughout this study. Among the 36 strains, strains SNA-1, SNA-3, and SNA-5 were isolated from samples respectively obtained by boring O-l, 1-3, and 3-5 cm in depth from the surface of corroded concrete at a sewage treatment plant in Saitama prefecture, and strains SYU-2 and SYU-4 from samples respectively obtained by boring O-2 and 2-4cm into corroded concrete obtained at a sewage treatment plant in Shizuoka prefecture. Strain OIT 5-10 was isolated from corroded concrete obtained at a sewage treatment plant in Okinawa prefecture. To isolate the bacteria, 1 g of corroded concrete was incubated at 30°C under aerobic conditions in an Fez+ medium (pH 2.5) containing FeS04. 7H@ (3%), (NHdzS04 (0.3%), K2HP04 (O-05%), MgS04.7H20 (0.05%), KC1 (O.Ol%), and Ca(NO& (0.001%). When the culture medium turned rusty, i.e., when the Fe2+ in the medium was oxidized by iron-oxidizing bacteria to Fe3+, aliquots of the culture medium were plated on gellan gum plates containing FeS04. 7HzO (2%), (NH&S04 (0.3%), K2HP04 (0.05%), MgS04.7H20 (0.05%), KC1 (O.Ol%), Ca(N03h
author. 300
VOL.
88,
1999
(O.OOl%), and yeast extract (0.03%). Rusty colonies appearing on the plate were picked up. This process was repeated more than three times and the final isolates were preserved on the Fe2+ medium (pH 2.5) and used throughout this study. Iron-oxidizing bacteria isolated were cultured in an elemental sulfur medium (pH 1.5) containing elemental sulfur (l.O%), (NH&S04 (0.3%), K2HP0., (0.05%), MgS04.7H20 (0.05%), KC1 (O.Ol%), and Ca(N03)2 (0.001%) or in an FeS medium (pH 1.5) containing FeS (1 .O%), (NH&S04 (0.3%), K2HPQ (0.05%), MgS04.7H20 (0.05%), KC1 (O.Ol%), and Ca(N03)2 (0.001%). Cells in the medium were counted Growth rate with a hemacytometer (Kayagaki Irika Kogyo, Tokyo) after dilution with 0.01 N H2S04 when necessary. Iron-oxidizing activity The iron-oxidizing activity of washed intact cells of iron-oxidizing bacteria was estimated from the oxygen uptake rate in a biological oxygen monitor (Yellow Springs Instrument Co., Yellow Springs, OH, USA). The reaction mixture was composed of 0.1 M P-alanine-SOa2- buffer, pH 3.0, (2.5 ml), washed intact cells, and FeS04. 7H20 (100 /*mol). The reaction was done at 30°C. The washed intact cells used were prepared as follows. Each strain of iron-oxidizing bacteria was grown in 20 1 of the Fe2+ medium (pH 1.5) under aeration at 28°C for a week. The culture medium was filtered with Toyo no. 2 filter paper to remove the bulk of the ferric precipitates and then centrifuged using a Hitachi 18PR-52 continuous-flow rotor at 15,OOOXg and a flow rate of 200mVmin. Harvested cells were washed three times with 0.1 M j3-alanine-OJ2- buffer @H 3.0). Fez+-dependent oxygen uptake activity of corroded concrete The Fe2+-dependent oxygen uptake activity
of corroded concrete was estimated in a Warburg manometer (Yanagimoto Mfg, Kyoto). A Warburg flask contained the reaction mixture (3.0 ml) in the reaction chamber and 20% potassium hydroxide (0.2ml) in the center well. The gas phase was air and the temperature was kept at 30°C. The reaction mixture in the Warburg flask was composed of 0.1 M p-alanine-SOd2- buffer, pH 3.0, (2.5 ml), corroded concrete (1.0 g), and FeS04. TABLE
IRON-OXIDIZING
IN CORRODED
CONCRETE
301
7H20 (100 pmol). To eliminate any endogenous oxygen uptake by microorganisms in the concrete sample other than iron-oxidizing bacteria, the oxygen uptake activity was also measured using the above reaction mixture without Fe2+. The reaction was started by adding Fe2+ to the reaction mixture and carried out at 30°C. Activity of hydrogen sulfide: ferric ion oxidoreductase The SFORase activity was determined by (SFORase)
measuring the Fe2+ produced in a reaction mixture under aerobic conditions in the presence of sodium cyanide, an inhibitor of iron oxidase (16). The reaction mixture was composed of 0.1 M j-alanine-S042p buffer, pH 3.0, (8.0ml), elemental sulfur (2OOmg), Fe3+ (50 pmol), washed intact cells, and sodium cyanide (50 pmol). The total volume was 10 ml. The reaction was carried out by shaking the mixture at 30°C. A sample was withdrawn and centrifuged at 12,OOOxg for 2 min to remove elemental sulfur and cells, and the concentration of Fe2+ produced in the supernatant was measured by the o-phenanthroline method (19). Activity of 14C02 uptake into washed intact cells
The carbon dioxide (C02) fixation activity was estimated from the amount of Na214C03 assimilated by the cells. The composition of the reaction mixture was as follows: 4.0 ml of 0.1 M ,B-alanine-S042p buffer, pH 3.0; washed intact cells of an iron-oxidizing bacterium grown in a Fe2+ medium (pH 2.5), 1 mg of protein; carrier Na2C03, 1 pmol; Na214C03, 1 ,&i; and FeSO,. 7H20 (20 pmol) or elemental sulfur (200mg). The total volume of the reaction mixture was 5.0ml. The reaction mixture without sodium carbonate and FeS04. 7H20 or elemental sulfur was incubated at 30°C for 10min. The reaction was started by adding Na214C03, carrier Na2C03, and FeS04. ‘7&O or elemental sulfur. The reaction mixture (0.55 ml) was withdrawn and then passed through a 0.45-pm membrane filter. The filter with cells was washed three times with 10ml water and then put into 4.Oml of a counting sol (Scintisol EX-H). After the filter had been completely solubilized in the counting mixture, the radioactivity was measured with an Aloka LSC-635 liquid scintillation system.
1. Growth of cells in elementalsulfur and FeS media, iron-oxidizing activity, and SFORase activity of iron-oxidizing isolated from sewage plant corroded concrete
Growth of cells in medium with elemental sulfur FeS ( x lo6 cells/ml/20 d) ( x lo6 cells/ml/20 d) Strains isolated from corroded concrete OIT 4-l 11 33 OIT 4-10 210 230 OIT 5-10 162 205 OMO 4- 1 52 37 OMO 4-2 48 11 01s 3-10 174 141 01s 3-100 96 93 ONA 4-l 224 125 HNI 4-l 16 9 HNI 4-10 148 80 SYU 3-2 250 92 SNA-1 51 42 SNA-3 288 120 SNA-5 88 33 NAB 3-1 70 27 Thiobacilius ferrooxidans AP19-3 152 58
Strain
BACTERIA
Iron-oxidizing activity Oz uptake (pl/mg/min) 63 60 85 54 56 149 120 84 132 86 279 246
bacteria
SFORase activity= (pmol/mg/min)
0.13 0.16 0.22 0.07 0.09 0.07 0.15 0.37 0.09 0.15 0.35 0.05 124 0.11 132 0.20 242 0.50 182 0.51 a The activity of hydrogen sulfide: ferric ion oxidoreductase (SFORase) (16) was determined by measuring the Fe2+ produced in the reaction mixture as described in Materials and Methods.
302
MAEDA
ET AL.
RESULTS Isolation of iron-oxidizing bacteria from corroded concrete Thirty-six strains of iron-oxidizing bacteria were isolated from corroded concrete samples obtained at eight sewage treatment plants in Japan, When grown statically in Fez+ medium (pH 1.5), the isolates gave cell yields ranging from 5-13 X lo* cells/ml. The strains grew autotrophically not only in Fe2+ medium but also in sulfur and FeS media. The growth of 15 representative strains in sulfur and FeS media is shown in Table 1. The growth yields in the two media differed from strain to strain. Two kinds of mesophilic iron-oxidizing bacteria are known, T. ferrooxidans and Leptospirillum ferrooxidans. T. ferrooxidans can use either ferrous iron or elemental sulfur as a sole source of energy, whereas L. ferrooxidans is able to use only ferrous iron. Our experimental results on energy source utilization thus suggest that the 36 isolates were T. ferrooxidans strains. Their iron-oxidizing activity ranged from 54 to 279 ,& mg/min (Table 1). T. ferrooxidans are known to possess hydrogen sulfide: ferric ion oxidoreductase (SFORase), which catalyzes the oxidation of elemental sulfur with ferric ion as an electron acceptor to give sulfite and ferrous ion (16, 19). SFORase plays a crucial role in sulfur oxidation by T. ferrooxidans. All the iron-oxidizing bacteria isolated from corroded concrete had SFORase activity, ranging from 0.05 to 0.50 /*mol/mg/min (Table l), which accords with their ability to grow in a sulfur medium. Electron micrographs of strains SNA-1, SNA-5, OIT 5-10, and T. ferrooxidans AP19-3 are shown in Fig. 1. Like a control strain T. ferrooxidans AP19-3 (16), the isolates were all strait rods (Fig. 1). Strain SNA-5 (0.50.6 x 0.9-1.5 pm) was gram-negative and non-spore-forming. The mean G-t C content of its DNA was 55.9 mol% (data not shown). Kuenen and Tuovinen reported the mean GS C content of T. ferrooxidans DNA to be 53.660.1 mol% (20), which is further evidence that strain SNA-5 is a T. ferrooxidans strain. SNA-5 and SNA-1 had optimum pH and temperature optima for growth of 1.5 and 30°C (data not shown), indicating that they are acidophilic and mesophilic iron-oxidizing bacteria. Carbon dioxide fixation activity The i4C02 fixation activities of strains SNA-5 and T. ferrooxidans AP19-3 were studied with washed intact cells grown in Fez+ medium (pH2.5). Strain SNA-5 as well as T. ferrooxidans AP19-3 assimilated 14C02 when ferrous iron or elemental sulfur was used as an energy source (Fig. 2). The specific activities of 14C02 fixation of strain SNA-5 and T. ferrooxidans AP19-3 were 56 and 171 nmol/mg/ 60 min, respectively. This result supports that strain SNA5 grew autotrophically in Fe2+, sulfur and FeS media. Estimating the number of iron-oxidizing bacteria in corroded concrete samples Since iron-oxidizing bacteria were isolated from all of the corroded concrete samples tested, we considered it important to estimate the numbers of bacteria in l-g samples taken at various depths in the concrete in order to better understand the role of iron-oxidizing bacteria in concrete corrosion. To our knowledge, this is the first time that such an estimate has been made. Samples were obtained from different depth by boring into the concrete structure at sewage treatment plants. Samples SNA (O-l cm) SNA (l-3 cm) and SNA (3-5 cm) were obtained from corroded concrete at a plant in Saitama prefecture at depths O-l, l-3 and 3-5 cm, respectively, below the concrete surface.
FIG. 1. Electron micrographs of iron-oxidizing bacteria isolated from corroded concrete and of T. firrooxidans AP19-3. (A) Strain SNA-1; (B) strain SNA-5; (C) strain OIT 5-10: (D) T. ferrooxirfms AP19-3. The bar represents 1.0 :rm.
The concentrations of oxygen, hydrogen sulfide, and carbon monoxide in the atmosphere above the sampling site were 17.7%, 96 ppm, and 1 ppm, respectively. The level of corrosion in sample SNA (O-l cm) was much more severe than in SNA (1-3 cm) and SNA (3-5 cm). The pHs of samples SNA (O-l cm), SNA (l-3 cm) and SNA (35 cm) were 1.4, 1.7, and 2.0, respectively (Table 2). The fact that high concentrations of hydrogen sulfide (96 ppm) and molecular oxygen (17.7%) were observed in the atmosphere above the sampling site suggests that the strong acid (pH 1.4) produced at the surface of the corroded concrete is highly likely to be sulfuric acid rather than hydrochloric, nitric, or carbonic acid, because it is known that T. thiooxidans and T. ferrooxidans can oxidize hydrogen sulfide with molecular oxygen as an electron acceptor to give sulfuric acid. Similar pH and corrosion level trends were observed in the corroded concrete samples obtained from a sewage treat-
VOL.
IRON-OXIDIZING
88. 1999
BACTERIA
IN CORRODED
CONCRETE
303
125 125 (A) (A) Sk
E
10
20 Time
30
40
8
0
(min)
50 Time
100 (min)
FIG. 2. 14C02 uptake activities of an iron-oxidizing bacterium isolated from corroded concrete and of T. ferrooxiduns AP19-3. The 14C02 uptake activities of washed intact cells of T. ferrooxidans AP19-3 ( n ) and strain SNA-5 (0) isolated from corroded concrete obtained from a sewage treatment plant in Saitama prefecture were measured in the presence of ferrous iron (A) or elemental sulfur (B) as an electron donor for the 14C0, assimilation reaction. The 14C0, uotake activities of washed intact cells of T. ferrooxiduns AP19-3 ( 0 ) and strain SNA-5 (0) were also measure: in the reaction mixture with 20mM HgC12.
ment plant in Shizuoka prefecture, SYU (O-2 cm) and SYU (2-4 cm), which had pHs of 1.4 and 1.7, respectively. To estimate the number of iron-oxidizing bacteria in 1 g of corroded concrete, we used, for the first time, a biological reaction specific to iron-oxidizing bacteria. T. ferrooxidans cells oxidizes Fe2+ under aerobic and acidic conditions as in the following equation: 4FeS04 + O2 + 10H20 ---,4Fe(OH)3 + 4H2S04 If 1 g of corroded concrete is added to an acidic reaction mixture containing Fe2+, iron-oxidizing bacteria in the concrete oxidize the Fe2+ with molecular oxygen as an electron acceptor. As a result, Fe2+-dependent oxygen uptake is observed, with the level of oxygen uptake activity being dependent on the number of iron-oxidizing bacteria in the sample. Thus, we first plotted an authentic curve to determine the relationship between the cell number and Fe2+-dependent iron-oxidizing activity for each of the strains isolated from the corroded concrete samples (Fig. 3). The iron-oxidizing activity per lo* cells of the iron-oxidizing bacteria was different from each strain. The activity of strain SNA-1 was approximately 4 times higher than that of strain SNA-3. We next measured the oxygen uptake of 1 g of each corroded concrete samples in the reaction mixture with Fez+ (Fig. 4). The oxygen uptake activity was also measured in the reaction mixture with 1 g of each samples but without Fe2+ to eliminate any oxygen uptake due to the endogenous respiration of heterotrophic microorganisms in the samples. The amount of oxygen taken up in the reaction mixture with Fez+ was higher than the uptake without TABLE
2.
Number of iron-oxidizing
Corroded concrete
Deptha
PH
SNA SNA SNA SYU SYU
O-l cm l-3 cm 3-5 cm O-2 cm 2-4 cm
1.4 1.7 2.0 1.4 1.7
(O-l (l-3 (3-5 (O-2 (2-4
cm)h cm)h cm)h crnp cm)
Fe2+. The difference in oxygen uptake activity between reaction mixtures with or without Fe2+ was considered to be Fe2+-dependent oxygen uptake activity caused by iron-oxidizing bacteria in the corroded concrete sample. Using this Fez+-dependent oxygen uptake activity (Fig. 4) and the previously determined relationship between the cell number and iron-oxidizing activity (Fig. 3), the number of iron-oxidizing bacterial cells in 1 g of the corroded concrete samples SNA (O-l cm), SNA (l-3 cm), SNA (3-5 cm), SYU (O-2cm), and SYU (2-4cm) was calculated to be 1.2x 108, 5 x 107, 5 x 106, 3.1 x lo7 and 2.0X lo7 cells, respectively (Table 2). Among the SNA samples, the one with the lowest pH contained many more iron-oxidizing bacterial cells than that whose pH was the highest. DISCUSSION
Iron-oxidizing bacteria were isolated from all the corroded concrete samples obtained at eight sewage treatment plants located in different regions of Japan. A close relationship was observed among three parameters: the degree of concrete corrosion, the number of ironoxidizing bacteria in the sample, and the level of acidity produced in the corroded concrete. The degree of concrete corrosion was shown to be much more severe at the surface of a concrete structure than in the interior. The pH of samples obtained at or near the surface of corroded concrete was lower than that of deeper samples, indicating that the level of acidification was higher at the surface than in interior. Since both hydrogen
bacteria in corroded concrete from sewage treatment plants Fez&-dependent O2 uptake activity (pmoI/min/g concrete) 0.67 0.07
0.04 0.08 0.05
a Depth from the surface of corroded concrete. b Corroded concrete samples from a sewage treatment plant in Saitama prefecture. c Corroded concrete samples from a sewage treatment plant in Shizuoka prefecture.
Number of cells in corroded concrete (X 10’ cells/g concrete) 12.0 5.0 0.5 3.1 2.0
304
MAEDA
ET AL.
.I. BIOSCX.BIOENG. tron acceptor, i.e., microbial sulfur specific to T. thiooxidans cells.
oxidation
is no1
REFERENCES
0
I
2
3
Number of cells in reaction mixture (x 10s cells) FIG. 3. Relationship between cell number and iron-oxidizing activity. The iron-oxidizing activities of T. ferrooxidans AP19-3 (A) and iron-oxidizing bacterial strains SNA-1 ( l ), SNA-3 (0), SNA-5 ( n ), SYU-2 ( 0 ), and SYU-4 (A) isolated from corroded concrete were measured using an oxygen electrode. The number of cells in the reaction mixture was counted with hemacytometer.
sulfide and oxygen were observed in the atmosphere at the sewage treatment plants from which the corroded concrete samples were obtained, our findings strongly suggest that an iron-oxidizing bacterium is involved in acid production from hydrogen sulfide generated in sewage treatment plants, and consequently, in concrete corrosion because concrete is mainly composed of calcium oxides, hydroxides, and carbonates, and these compounds can react with sulfuric acid to form calcium sulfate (gypsum), thereby reducing the strength of the concrete structure (4). The findings that the pH was lower and the number of iron-oxidizing bacteria was higher at the concrete surface than in the interior seem reasonable given that the iron-oxidizing bacterium T. ferrooxidans is an aerobic and acidophilic chemolithotroph which produces much more sulfuric acid at sites where hydrogen sulfide and molecular oxygen are more abundant. Unfortunately, we were unable to estimate the cell number of the sulfur-oxidizing bacterium T. thiooxidans in 1 g of corroded concrete from the sulfur-dependent oxygen uptake activity because T. ferrooxidans and T. acidophilus (21) also have the ability to oxidize elemental sulfur at pH 2-3 with molecular oxygen as an elec-
Time (min)
1. Parker, C. D.: The corrosion of concrete, I. The isolation of a species of bacterium associated with the corrosion of concrete exposed to atmospheres containing hydrogen sulphide. Aust. .J. Exp. Biol. Med. Sci., 23, 81-90 (1945). 2. Parker, C. D.: The corrosion of concrete, Ii. The function of Thiobacillus concretivorus (nov. spec.) in the corrosion of concrete exposed to atmospheres containing hydrogen sulphide. Aust. J. Exp. Biol. Med. Sci., 23, 91-98 (1945). 3. Parker, C. D.: Species of sulphur bacteria associated with the corrosion of concrete. Nature, 159, 439-440 (1947). 4. Sand, W.: Importance of hydrogen sulfide, thiosulfate, and methylmercaptan for growth of thiobacilli during simulation of concrete corrosion. Appl. Environ. Microbial., 53, 1645-1648 (1987). 5. Maeda, T., Negishi, A., Nogami, Y., and Sugio, T.; Nickel inhibition of the growth of a sulfur-oxidizing bacterium isolated from corroded concrete. Biosci. Biotech. Biochem., 60, 626629 (1996). 6. Nogami, Y., Maeda, T., Negishi, A., and Sugio, T.: Inhibition of sulfur oxidizing activity by nickel ion in Thiobacillus thiooxidans NBl-3 isolated from the corroded concrete. Biosci. Biotech. Biochem., 61, 1373-1375 (1996). I. Milde, K., Sand, W., Wolff, W., and Bock, E.: Thiobacilli of the corroded concrete walls of the Hamburg sewer system. J. Gen. Microbial., 129, 1327-1333 (1983). 8. Sand, W. and Bock, E.: Concrete corrosion in the Hamburg sewer system. Environ. Tech. Lett., 5, 517-528 (1984). 9. Yoshida, N., Morinaga, T., and Murooka, Y.: Characterization and identification of bacterial strains isolated from corroded concrete in the accumulation stratum and their resistance levels to heavy metals. J. Ferment. Bioeng., 76, 400-402 (1993). 10. Maeda, T., Negishi, A., Oshima, Y., Nogami, Y., Kamimura, K., and Sugio, T.: Isolation of a sulfur-oxidizing bacterium that can grow under alkaline pH, from corroded concrete. Biosci. Biotech. Biochem., 62, 1087-1092 (1998). 11 Torma, A. E.: The role of Thiobacillus ferrooxidans in hydrometallurgical processes. Adv. Biochem. Eng., 6, l-38 (1977). 12. Lundgren, Dl G. and Silver, M.: Ore leaching by bacteria. Ann. Rev. Microbial., 34, 263-283 (1980). 13. Hutchins, S. R., Davidson, M. S., Brierley, J. A., and Brierley, C. I,.: Microorganisms in reclamation of metals. Ann. Rev. Microbial., 40, 311-336 (1986). 14. Brierley, C. L.: Microbiological mining. Sci. Am., 247, 42-49 (1982). 15. Nicolaidis, A. A.: Microbial mineral processing: the opportunities for genetic manipulation. J. Chem. Tech. Biotechnol., 38, A..
Time (min)
Time (min)
FIG. 4. Fe2+-dependent oxygen uptake activities of corroded concrete samples. Iron-oxidizing bacteria in corroded concrete samples oxidized Fe2+ in the reaction mixture and thus showed Fe2+-dependent oxygen uptake activity. The oxygen uptake activities for 1 g of corroded concrete samples (A) SNA (O-l cm), (B) SNA (l-3 cm), and (C) SNA (3-5 cm) were measured with (0) or without (0) 100 /*mol of ferrous iron. Samples SNA (O-l cm), SNA (l-3 cm), and SNA (3-5 cm) were obtained by boring O-l, l-3, and 3-5 cm in depth from the surface of corroded concrete at a sewage treatment plant in Saitama prefecture.
VOL.
88, 1999
167-185 (1987). 16. Sugio, T., Mizunashi, W., Inagaki, K., and Taoo, T.: Purification and some properties of sulfur: ferric ion oxidoreductase from Thiobacillus ferrooxidans. J. Bacterial., 169, 4916-4922 (1987). 17. Sugio, T., Hirose, T., Ye Li-zhen, and Tano, T.: Purification and some properties of sulfite: ferric ion oxidoreductase from Thiobuciilus ferrooxidans. J. Bacterial., 174, 4189-4192 (1992). 18. Sugio, T., Kanao, T., Furukawa, H., Nagasawa, T., and Blake, R. C: Isolation and identification of an iron-oxidizing bacterium which can grow on tetrathionate medium and the properties of tetrathionate-decomposing enzyme isolated from the bac-
IRON-OXIDIZING
BACTERIA
IN CORRODED
CONCRETE
305
terium. J. Ferment. Bioeng., 82, 233-238 (1996). 19. Sugio, T., Domatsu, T., Munakata, O., Tano, T., and Imai, K.: Role of a ferric ion-reducing system in sulfur oxidation of Thiobacillus ferrooxidans. Appl. Environ. Microbial., 49, 1401-1406 (1985). 20. Kuenen, J. G. and Tuovinen, 0. H.: The Genera Thiobacillus and Thiomicrospira, p. 1023-1036. In Starr, M. P., Stolp, H., Truper. H. G.. Balows. A.. and Schleael. H. G. (ed.). The prokaryotes. Springer-V&lag; Berlin (198;)’ . ” 21. Guay, R. and Silver, M.: Thiobacillus acidophilus sp. nov.; isolation and some physiological characteristics. Can. J. Microbiol., 21, 281-288 (1974).
Isolation of Iron-Oxidizing Bacteria from Corroded Concretes of SewageTreatment Plants TERUNOBU
MAEDA,’ ATSUNORI NEGISHI,’ HIROMI KOMOT0,2 YUKO OSHIMA,2 KAZUO KAMIMURA,2 AND TSUYOSHI SUGI02* Technical Research Institute, Hazama Corporation, 515-l Nishimukai, Karima, Tsukuba 305-0822’ and Department of Biological Function and Genetic Resources Science, Faculty of Agriculture, Okayama University, I-I-I Tsushima Naka, Okayama 700-8530,2 Japan Received 10 March 1999/Accepted 10 June 1999
Thirty-six strains of iron-oxidizing bacteria were isolated from corroded concrete samples obtained at eight sewage treatment plants in Japan. All of the strains isolated grew autotrophically in ferrous sulfate (3.0%), elemental sulfur (1.0%) and FeS (1.0%) media (pH 1.5). Washed intact cells of the 36 isolates had activities to oxidize both ferrous iron and elemental sulfur. Strain SNA-5, a representative of the isolated strains, was a gram-negative, rod-shaped bacterium (OS-o.6 X 0.9-1.5 q). The mean G+C content of its DNA was 55.9 mol%. The pH and temperature optima for growth were 1.5 and 30% and the bacterium had activity to assimilate ‘4c02 into the cells when ferrous iron or elemental sulfur was used as a sole source of energy. These results suggest that SNAQ is Thiobacillus ferrooxidans strain. The pHs and numbers of iron-oxidizing bacteria in corroded concrete samples obtained by boring to depths of O-l, 1-3, and 3-5 cm below the concrete surface were respectively 1.4, 1.7, and 2.0, and 1.2 x 108, 5 X lo’, and 5 X lo6 cells/g concrete. The degree of corrosion in the sample obtained nearest to the surface was more severe than in the deeper samples. The findings indicated that the levels of acidification and corrosion of the concrete structure corresponded with the number of iron-oxidizing bacteria in a concrete sample. Sulfuric acid produced by the chemolithoautotrophic sulfuroxidizing bacterium Thiobacillus thiooxidansis known to induce concrete corrosion. Since not only T. thiooxidans but also T. ferrooxidans can oxidize reduced sulfur compounds and produce sulfuric acid, the results strongly suggest that T. ferrooxidans as well as T. thiooxidans is involved in concrete corrosion. [Key words: corrosion]
iron-oxidizing
bacterium,
ThiobaciNus ferrooxidans, sulfur-oxidizing
Parker first presented evidence that concrete structures are corroded by sulfuric acid produced by the sulfuroxidizing bacterium Thiobacillus concretivourus (l-3). Concrete in its main structure is composed of calcium oxides, hydroxides, and carbonates, and these compounds can react with sulfuric acid to give calcium sulfate (gypsum) (4). This resultant calcium loss greatly diminishes the strength of concrete structures. The bacterium isolated by Parker was later identified as Thiobacillus thiooxidans, and thiobacilli other than T. thiooxidans (5, 6), namely Thiobacillus neapolitanus, T. in termed&, Thiobacillus novellus, and Thiobacillus versutus have also been isolated from sewer systems and corroded concrete (7-10). The iron-oxidizing bacterium Thiobacillus ferrooxidans mainly inhabits drainage areas in acid mines and plays a crucial role in the bacterial leaching of sulfide ores (11-15). As well as T. thiooxidans, T. ferrooxidans is also known to oxidize reduced inorganic sulfur compounds and produce sulfuric acid. The enzymes involved in the oxidation of inorganic sulfur compounds have been purified from the bacterium (16-19). The ability of T. ferrooxidans to produce sulfuric acid suggested that in addition to T. thiooxidans, T. ferrooxidans strains are also likely to be present in corroded concrete and play a role in concrete corrosion. We were able to confirm that this is the case, as well as to clarify the mechanism of concrete corrosion more precisely, by isolating iron-oxidizing bacteria from corroded concrete samples obtained from eight sewage treatment plants in * Corresponding
bacterium,
concrete,
Japan. To our knowledge, there have been no previous reports on concrete corrosion by microorganisms that have focussed on iron-oxidizing bacteria. MATERIALS
AND METHODS
Microorganisms, media, conditions of cultivation, and isolation procedures T. ferrooxidans AP19-3 (16, 19) and 36 strains of iron-oxidizing bacteria isolated from corroded concrete samples obtained at eight sewage treatment plants in Japan were used throughout this study. Among the 36 strains, strains SNA-1, SNA-3, and SNA-5 were isolated from samples respectively obtained by boring O-l, 1-3, and 3-5 cm in depth from the surface of corroded concrete at a sewage treatment plant in Saitama prefecture, and strains SYU-2 and SYU-4 from samples respectively obtained by boring O-2 and 2-4cm into corroded concrete obtained at a sewage treatment plant in Shizuoka prefecture. Strain OIT 5-10 was isolated from corroded concrete obtained at a sewage treatment plant in Okinawa prefecture. To isolate the bacteria, 1 g of corroded concrete was incubated at 30°C under aerobic conditions in an Fez+ medium (pH 2.5) containing FeS04. 7H@ (3%), (NHdzS04 (0.3%), K2HP04 (O-05%), MgS04.7H20 (0.05%), KC1 (O.Ol%), and Ca(NO& (0.001%). When the culture medium turned rusty, i.e., when the Fe2+ in the medium was oxidized by iron-oxidizing bacteria to Fe3+, aliquots of the culture medium were plated on gellan gum plates containing FeS04. 7HzO (2%), (NH&S04 (0.3%), K2HP04 (0.05%), MgS04.7H20 (0.05%), KC1 (O.Ol%), Ca(N03h
author. 300
VOL.
88,
1999
(O.OOl%), and yeast extract (0.03%). Rusty colonies appearing on the plate were picked up. This process was repeated more than three times and the final isolates were preserved on the Fe2+ medium (pH 2.5) and used throughout this study. Iron-oxidizing bacteria isolated were cultured in an elemental sulfur medium (pH 1.5) containing elemental sulfur (l.O%), (NH&S04 (0.3%), K2HP0., (0.05%), MgS04.7H20 (0.05%), KC1 (O.Ol%), and Ca(N03)2 (0.001%) or in an FeS medium (pH 1.5) containing FeS (1 .O%), (NH&S04 (0.3%), K2HPQ (0.05%), MgS04.7H20 (0.05%), KC1 (O.Ol%), and Ca(N03)2 (0.001%). Cells in the medium were counted Growth rate with a hemacytometer (Kayagaki Irika Kogyo, Tokyo) after dilution with 0.01 N H2S04 when necessary. Iron-oxidizing activity The iron-oxidizing activity of washed intact cells of iron-oxidizing bacteria was estimated from the oxygen uptake rate in a biological oxygen monitor (Yellow Springs Instrument Co., Yellow Springs, OH, USA). The reaction mixture was composed of 0.1 M P-alanine-SOa2- buffer, pH 3.0, (2.5 ml), washed intact cells, and FeS04. 7H20 (100 /*mol). The reaction was done at 30°C. The washed intact cells used were prepared as follows. Each strain of iron-oxidizing bacteria was grown in 20 1 of the Fe2+ medium (pH 1.5) under aeration at 28°C for a week. The culture medium was filtered with Toyo no. 2 filter paper to remove the bulk of the ferric precipitates and then centrifuged using a Hitachi 18PR-52 continuous-flow rotor at 15,OOOXg and a flow rate of 200mVmin. Harvested cells were washed three times with 0.1 M j3-alanine-OJ2- buffer @H 3.0). Fez+-dependent oxygen uptake activity of corroded concrete The Fe2+-dependent oxygen uptake activity
of corroded concrete was estimated in a Warburg manometer (Yanagimoto Mfg, Kyoto). A Warburg flask contained the reaction mixture (3.0 ml) in the reaction chamber and 20% potassium hydroxide (0.2ml) in the center well. The gas phase was air and the temperature was kept at 30°C. The reaction mixture in the Warburg flask was composed of 0.1 M p-alanine-SOd2- buffer, pH 3.0, (2.5 ml), corroded concrete (1.0 g), and FeS04. TABLE
IRON-OXIDIZING
IN CORRODED
CONCRETE
301
7H20 (100 pmol). To eliminate any endogenous oxygen uptake by microorganisms in the concrete sample other than iron-oxidizing bacteria, the oxygen uptake activity was also measured using the above reaction mixture without Fe2+. The reaction was started by adding Fe2+ to the reaction mixture and carried out at 30°C. Activity of hydrogen sulfide: ferric ion oxidoreductase The SFORase activity was determined by (SFORase)
measuring the Fe2+ produced in a reaction mixture under aerobic conditions in the presence of sodium cyanide, an inhibitor of iron oxidase (16). The reaction mixture was composed of 0.1 M j-alanine-S042p buffer, pH 3.0, (8.0ml), elemental sulfur (2OOmg), Fe3+ (50 pmol), washed intact cells, and sodium cyanide (50 pmol). The total volume was 10 ml. The reaction was carried out by shaking the mixture at 30°C. A sample was withdrawn and centrifuged at 12,OOOxg for 2 min to remove elemental sulfur and cells, and the concentration of Fe2+ produced in the supernatant was measured by the o-phenanthroline method (19). Activity of 14C02 uptake into washed intact cells
The carbon dioxide (C02) fixation activity was estimated from the amount of Na214C03 assimilated by the cells. The composition of the reaction mixture was as follows: 4.0 ml of 0.1 M ,B-alanine-S042p buffer, pH 3.0; washed intact cells of an iron-oxidizing bacterium grown in a Fe2+ medium (pH 2.5), 1 mg of protein; carrier Na2C03, 1 pmol; Na214C03, 1 ,&i; and FeSO,. 7H20 (20 pmol) or elemental sulfur (200mg). The total volume of the reaction mixture was 5.0ml. The reaction mixture without sodium carbonate and FeS04. 7H20 or elemental sulfur was incubated at 30°C for 10min. The reaction was started by adding Na214C03, carrier Na2C03, and FeS04. ‘7&O or elemental sulfur. The reaction mixture (0.55 ml) was withdrawn and then passed through a 0.45-pm membrane filter. The filter with cells was washed three times with 10ml water and then put into 4.Oml of a counting sol (Scintisol EX-H). After the filter had been completely solubilized in the counting mixture, the radioactivity was measured with an Aloka LSC-635 liquid scintillation system.
1. Growth of cells in elementalsulfur and FeS media, iron-oxidizing activity, and SFORase activity of iron-oxidizing isolated from sewage plant corroded concrete
Growth of cells in medium with elemental sulfur FeS ( x lo6 cells/ml/20 d) ( x lo6 cells/ml/20 d) Strains isolated from corroded concrete OIT 4-l 11 33 OIT 4-10 210 230 OIT 5-10 162 205 OMO 4- 1 52 37 OMO 4-2 48 11 01s 3-10 174 141 01s 3-100 96 93 ONA 4-l 224 125 HNI 4-l 16 9 HNI 4-10 148 80 SYU 3-2 250 92 SNA-1 51 42 SNA-3 288 120 SNA-5 88 33 NAB 3-1 70 27 Thiobacilius ferrooxidans AP19-3 152 58
Strain
BACTERIA
Iron-oxidizing activity Oz uptake (pl/mg/min) 63 60 85 54 56 149 120 84 132 86 279 246
bacteria
SFORase activity= (pmol/mg/min)
0.13 0.16 0.22 0.07 0.09 0.07 0.15 0.37 0.09 0.15 0.35 0.05 124 0.11 132 0.20 242 0.50 182 0.51 a The activity of hydrogen sulfide: ferric ion oxidoreductase (SFORase) (16) was determined by measuring the Fe2+ produced in the reaction mixture as described in Materials and Methods.
302
MAEDA
ET AL.
RESULTS Isolation of iron-oxidizing bacteria from corroded concrete Thirty-six strains of iron-oxidizing bacteria were isolated from corroded concrete samples obtained at eight sewage treatment plants in Japan, When grown statically in Fez+ medium (pH 1.5), the isolates gave cell yields ranging from 5-13 X lo* cells/ml. The strains grew autotrophically not only in Fe2+ medium but also in sulfur and FeS media. The growth of 15 representative strains in sulfur and FeS media is shown in Table 1. The growth yields in the two media differed from strain to strain. Two kinds of mesophilic iron-oxidizing bacteria are known, T. ferrooxidans and Leptospirillum ferrooxidans. T. ferrooxidans can use either ferrous iron or elemental sulfur as a sole source of energy, whereas L. ferrooxidans is able to use only ferrous iron. Our experimental results on energy source utilization thus suggest that the 36 isolates were T. ferrooxidans strains. Their iron-oxidizing activity ranged from 54 to 279 ,& mg/min (Table 1). T. ferrooxidans are known to possess hydrogen sulfide: ferric ion oxidoreductase (SFORase), which catalyzes the oxidation of elemental sulfur with ferric ion as an electron acceptor to give sulfite and ferrous ion (16, 19). SFORase plays a crucial role in sulfur oxidation by T. ferrooxidans. All the iron-oxidizing bacteria isolated from corroded concrete had SFORase activity, ranging from 0.05 to 0.50 /*mol/mg/min (Table l), which accords with their ability to grow in a sulfur medium. Electron micrographs of strains SNA-1, SNA-5, OIT 5-10, and T. ferrooxidans AP19-3 are shown in Fig. 1. Like a control strain T. ferrooxidans AP19-3 (16), the isolates were all strait rods (Fig. 1). Strain SNA-5 (0.50.6 x 0.9-1.5 pm) was gram-negative and non-spore-forming. The mean G-t C content of its DNA was 55.9 mol% (data not shown). Kuenen and Tuovinen reported the mean GS C content of T. ferrooxidans DNA to be 53.660.1 mol% (20), which is further evidence that strain SNA-5 is a T. ferrooxidans strain. SNA-5 and SNA-1 had optimum pH and temperature optima for growth of 1.5 and 30°C (data not shown), indicating that they are acidophilic and mesophilic iron-oxidizing bacteria. Carbon dioxide fixation activity The i4C02 fixation activities of strains SNA-5 and T. ferrooxidans AP19-3 were studied with washed intact cells grown in Fez+ medium (pH2.5). Strain SNA-5 as well as T. ferrooxidans AP19-3 assimilated 14C02 when ferrous iron or elemental sulfur was used as an energy source (Fig. 2). The specific activities of 14C02 fixation of strain SNA-5 and T. ferrooxidans AP19-3 were 56 and 171 nmol/mg/ 60 min, respectively. This result supports that strain SNA5 grew autotrophically in Fe2+, sulfur and FeS media. Estimating the number of iron-oxidizing bacteria in corroded concrete samples Since iron-oxidizing bacteria were isolated from all of the corroded concrete samples tested, we considered it important to estimate the numbers of bacteria in l-g samples taken at various depths in the concrete in order to better understand the role of iron-oxidizing bacteria in concrete corrosion. To our knowledge, this is the first time that such an estimate has been made. Samples were obtained from different depth by boring into the concrete structure at sewage treatment plants. Samples SNA (O-l cm) SNA (l-3 cm) and SNA (3-5 cm) were obtained from corroded concrete at a plant in Saitama prefecture at depths O-l, l-3 and 3-5 cm, respectively, below the concrete surface.
FIG. 1. Electron micrographs of iron-oxidizing bacteria isolated from corroded concrete and of T. firrooxidans AP19-3. (A) Strain SNA-1; (B) strain SNA-5; (C) strain OIT 5-10: (D) T. ferrooxirfms AP19-3. The bar represents 1.0 :rm.
The concentrations of oxygen, hydrogen sulfide, and carbon monoxide in the atmosphere above the sampling site were 17.7%, 96 ppm, and 1 ppm, respectively. The level of corrosion in sample SNA (O-l cm) was much more severe than in SNA (1-3 cm) and SNA (3-5 cm). The pHs of samples SNA (O-l cm), SNA (l-3 cm) and SNA (35 cm) were 1.4, 1.7, and 2.0, respectively (Table 2). The fact that high concentrations of hydrogen sulfide (96 ppm) and molecular oxygen (17.7%) were observed in the atmosphere above the sampling site suggests that the strong acid (pH 1.4) produced at the surface of the corroded concrete is highly likely to be sulfuric acid rather than hydrochloric, nitric, or carbonic acid, because it is known that T. thiooxidans and T. ferrooxidans can oxidize hydrogen sulfide with molecular oxygen as an electron acceptor to give sulfuric acid. Similar pH and corrosion level trends were observed in the corroded concrete samples obtained from a sewage treat-
VOL.
IRON-OXIDIZING
88. 1999
BACTERIA
IN CORRODED
CONCRETE
303
125 125 (A) (A) Sk
E
10
20 Time
30
40
8
0
(min)
50 Time
100 (min)
FIG. 2. 14C02 uptake activities of an iron-oxidizing bacterium isolated from corroded concrete and of T. ferrooxiduns AP19-3. The 14C02 uptake activities of washed intact cells of T. ferrooxidans AP19-3 ( n ) and strain SNA-5 (0) isolated from corroded concrete obtained from a sewage treatment plant in Saitama prefecture were measured in the presence of ferrous iron (A) or elemental sulfur (B) as an electron donor for the 14C0, assimilation reaction. The 14C0, uotake activities of washed intact cells of T. ferrooxiduns AP19-3 ( 0 ) and strain SNA-5 (0) were also measure: in the reaction mixture with 20mM HgC12.
ment plant in Shizuoka prefecture, SYU (O-2 cm) and SYU (2-4 cm), which had pHs of 1.4 and 1.7, respectively. To estimate the number of iron-oxidizing bacteria in 1 g of corroded concrete, we used, for the first time, a biological reaction specific to iron-oxidizing bacteria. T. ferrooxidans cells oxidizes Fe2+ under aerobic and acidic conditions as in the following equation: 4FeS04 + O2 + 10H20 ---,4Fe(OH)3 + 4H2S04 If 1 g of corroded concrete is added to an acidic reaction mixture containing Fe2+, iron-oxidizing bacteria in the concrete oxidize the Fe2+ with molecular oxygen as an electron acceptor. As a result, Fe2+-dependent oxygen uptake is observed, with the level of oxygen uptake activity being dependent on the number of iron-oxidizing bacteria in the sample. Thus, we first plotted an authentic curve to determine the relationship between the cell number and Fe2+-dependent iron-oxidizing activity for each of the strains isolated from the corroded concrete samples (Fig. 3). The iron-oxidizing activity per lo* cells of the iron-oxidizing bacteria was different from each strain. The activity of strain SNA-1 was approximately 4 times higher than that of strain SNA-3. We next measured the oxygen uptake of 1 g of each corroded concrete samples in the reaction mixture with Fez+ (Fig. 4). The oxygen uptake activity was also measured in the reaction mixture with 1 g of each samples but without Fe2+ to eliminate any oxygen uptake due to the endogenous respiration of heterotrophic microorganisms in the samples. The amount of oxygen taken up in the reaction mixture with Fez+ was higher than the uptake without TABLE
2.
Number of iron-oxidizing
Corroded concrete
Deptha
PH
SNA SNA SNA SYU SYU
O-l cm l-3 cm 3-5 cm O-2 cm 2-4 cm
1.4 1.7 2.0 1.4 1.7
(O-l (l-3 (3-5 (O-2 (2-4
cm)h cm)h cm)h crnp cm)
Fe2+. The difference in oxygen uptake activity between reaction mixtures with or without Fe2+ was considered to be Fe2+-dependent oxygen uptake activity caused by iron-oxidizing bacteria in the corroded concrete sample. Using this Fez+-dependent oxygen uptake activity (Fig. 4) and the previously determined relationship between the cell number and iron-oxidizing activity (Fig. 3), the number of iron-oxidizing bacterial cells in 1 g of the corroded concrete samples SNA (O-l cm), SNA (l-3 cm), SNA (3-5 cm), SYU (O-2cm), and SYU (2-4cm) was calculated to be 1.2x 108, 5 x 107, 5 x 106, 3.1 x lo7 and 2.0X lo7 cells, respectively (Table 2). Among the SNA samples, the one with the lowest pH contained many more iron-oxidizing bacterial cells than that whose pH was the highest. DISCUSSION
Iron-oxidizing bacteria were isolated from all the corroded concrete samples obtained at eight sewage treatment plants located in different regions of Japan. A close relationship was observed among three parameters: the degree of concrete corrosion, the number of ironoxidizing bacteria in the sample, and the level of acidity produced in the corroded concrete. The degree of concrete corrosion was shown to be much more severe at the surface of a concrete structure than in the interior. The pH of samples obtained at or near the surface of corroded concrete was lower than that of deeper samples, indicating that the level of acidification was higher at the surface than in interior. Since both hydrogen
bacteria in corroded concrete from sewage treatment plants Fez&-dependent O2 uptake activity (pmoI/min/g concrete) 0.67 0.07
0.04 0.08 0.05
a Depth from the surface of corroded concrete. b Corroded concrete samples from a sewage treatment plant in Saitama prefecture. c Corroded concrete samples from a sewage treatment plant in Shizuoka prefecture.
Number of cells in corroded concrete (X 10’ cells/g concrete) 12.0 5.0 0.5 3.1 2.0
304
MAEDA
ET AL.
.I. BIOSCX.BIOENG. tron acceptor, i.e., microbial sulfur specific to T. thiooxidans cells.
oxidation
is no1
REFERENCES
0
I
2
3
Number of cells in reaction mixture (x 10s cells) FIG. 3. Relationship between cell number and iron-oxidizing activity. The iron-oxidizing activities of T. ferrooxidans AP19-3 (A) and iron-oxidizing bacterial strains SNA-1 ( l ), SNA-3 (0), SNA-5 ( n ), SYU-2 ( 0 ), and SYU-4 (A) isolated from corroded concrete were measured using an oxygen electrode. The number of cells in the reaction mixture was counted with hemacytometer.
sulfide and oxygen were observed in the atmosphere at the sewage treatment plants from which the corroded concrete samples were obtained, our findings strongly suggest that an iron-oxidizing bacterium is involved in acid production from hydrogen sulfide generated in sewage treatment plants, and consequently, in concrete corrosion because concrete is mainly composed of calcium oxides, hydroxides, and carbonates, and these compounds can react with sulfuric acid to form calcium sulfate (gypsum), thereby reducing the strength of the concrete structure (4). The findings that the pH was lower and the number of iron-oxidizing bacteria was higher at the concrete surface than in the interior seem reasonable given that the iron-oxidizing bacterium T. ferrooxidans is an aerobic and acidophilic chemolithotroph which produces much more sulfuric acid at sites where hydrogen sulfide and molecular oxygen are more abundant. Unfortunately, we were unable to estimate the cell number of the sulfur-oxidizing bacterium T. thiooxidans in 1 g of corroded concrete from the sulfur-dependent oxygen uptake activity because T. ferrooxidans and T. acidophilus (21) also have the ability to oxidize elemental sulfur at pH 2-3 with molecular oxygen as an elec-
Time (min)
1. Parker, C. D.: The corrosion of concrete, I. The isolation of a species of bacterium associated with the corrosion of concrete exposed to atmospheres containing hydrogen sulphide. Aust. .J. Exp. Biol. Med. Sci., 23, 81-90 (1945). 2. Parker, C. D.: The corrosion of concrete, Ii. The function of Thiobacillus concretivorus (nov. spec.) in the corrosion of concrete exposed to atmospheres containing hydrogen sulphide. Aust. J. Exp. Biol. Med. Sci., 23, 91-98 (1945). 3. Parker, C. D.: Species of sulphur bacteria associated with the corrosion of concrete. Nature, 159, 439-440 (1947). 4. Sand, W.: Importance of hydrogen sulfide, thiosulfate, and methylmercaptan for growth of thiobacilli during simulation of concrete corrosion. Appl. Environ. Microbial., 53, 1645-1648 (1987). 5. Maeda, T., Negishi, A., Nogami, Y., and Sugio, T.; Nickel inhibition of the growth of a sulfur-oxidizing bacterium isolated from corroded concrete. Biosci. Biotech. Biochem., 60, 626629 (1996). 6. Nogami, Y., Maeda, T., Negishi, A., and Sugio, T.: Inhibition of sulfur oxidizing activity by nickel ion in Thiobacillus thiooxidans NBl-3 isolated from the corroded concrete. Biosci. Biotech. Biochem., 61, 1373-1375 (1996). I. Milde, K., Sand, W., Wolff, W., and Bock, E.: Thiobacilli of the corroded concrete walls of the Hamburg sewer system. J. Gen. Microbial., 129, 1327-1333 (1983). 8. Sand, W. and Bock, E.: Concrete corrosion in the Hamburg sewer system. Environ. Tech. Lett., 5, 517-528 (1984). 9. Yoshida, N., Morinaga, T., and Murooka, Y.: Characterization and identification of bacterial strains isolated from corroded concrete in the accumulation stratum and their resistance levels to heavy metals. J. Ferment. Bioeng., 76, 400-402 (1993). 10. Maeda, T., Negishi, A., Oshima, Y., Nogami, Y., Kamimura, K., and Sugio, T.: Isolation of a sulfur-oxidizing bacterium that can grow under alkaline pH, from corroded concrete. Biosci. Biotech. Biochem., 62, 1087-1092 (1998). 11 Torma, A. E.: The role of Thiobacillus ferrooxidans in hydrometallurgical processes. Adv. Biochem. Eng., 6, l-38 (1977). 12. Lundgren, Dl G. and Silver, M.: Ore leaching by bacteria. Ann. Rev. Microbial., 34, 263-283 (1980). 13. Hutchins, S. R., Davidson, M. S., Brierley, J. A., and Brierley, C. I,.: Microorganisms in reclamation of metals. Ann. Rev. Microbial., 40, 311-336 (1986). 14. Brierley, C. L.: Microbiological mining. Sci. Am., 247, 42-49 (1982). 15. Nicolaidis, A. A.: Microbial mineral processing: the opportunities for genetic manipulation. J. Chem. Tech. Biotechnol., 38, A..
Time (min)
Time (min)
FIG. 4. Fe2+-dependent oxygen uptake activities of corroded concrete samples. Iron-oxidizing bacteria in corroded concrete samples oxidized Fe2+ in the reaction mixture and thus showed Fe2+-dependent oxygen uptake activity. The oxygen uptake activities for 1 g of corroded concrete samples (A) SNA (O-l cm), (B) SNA (l-3 cm), and (C) SNA (3-5 cm) were measured with (0) or without (0) 100 /*mol of ferrous iron. Samples SNA (O-l cm), SNA (l-3 cm), and SNA (3-5 cm) were obtained by boring O-l, l-3, and 3-5 cm in depth from the surface of corroded concrete at a sewage treatment plant in Saitama prefecture.
VOL.
88, 1999
167-185 (1987). 16. Sugio, T., Mizunashi, W., Inagaki, K., and Taoo, T.: Purification and some properties of sulfur: ferric ion oxidoreductase from Thiobacillus ferrooxidans. J. Bacterial., 169, 4916-4922 (1987). 17. Sugio, T., Hirose, T., Ye Li-zhen, and Tano, T.: Purification and some properties of sulfite: ferric ion oxidoreductase from Thiobuciilus ferrooxidans. J. Bacterial., 174, 4189-4192 (1992). 18. Sugio, T., Kanao, T., Furukawa, H., Nagasawa, T., and Blake, R. C: Isolation and identification of an iron-oxidizing bacterium which can grow on tetrathionate medium and the properties of tetrathionate-decomposing enzyme isolated from the bac-
IRON-OXIDIZING
BACTERIA
IN CORRODED
CONCRETE
305
terium. J. Ferment. Bioeng., 82, 233-238 (1996). 19. Sugio, T., Domatsu, T., Munakata, O., Tano, T., and Imai, K.: Role of a ferric ion-reducing system in sulfur oxidation of Thiobacillus ferrooxidans. Appl. Environ. Microbial., 49, 1401-1406 (1985). 20. Kuenen, J. G. and Tuovinen, 0. H.: The Genera Thiobacillus and Thiomicrospira, p. 1023-1036. In Starr, M. P., Stolp, H., Truper. H. G.. Balows. A.. and Schleael. H. G. (ed.). The prokaryotes. Springer-V&lag; Berlin (198;)’ . ” 21. Guay, R. and Silver, M.: Thiobacillus acidophilus sp. nov.; isolation and some physiological characteristics. Can. J. Microbiol., 21, 281-288 (1974).